MapikaJared PalmerDecision matchup
Decider 0.8Bvs.Kev 0.8B
Which model makes better decisions?
Decider 0.8B and Kev 0.8B perform about the same in independent tests.
Pick Decider 0.8B for
Pick Kev 0.8B for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
Zero-shot IE benchmark
levelDecider 0.8B is clearly ahead in 0 of 4 subtests, Kev 0.8B in none. 4 are level.
Zero-shot IE spectrum: mixed-pool classification accuracy
48 tasks, as of October 17, 2026
Zero-shot IE multilingual: medium-resource-language tier accuracy
18 tasks, as of October 17, 2026
Zero-shot IE multilingual: popular-language tier accuracy
18 tasks, as of October 17, 2026
Show all 4 subtests
| Subtest | Decider 0.8B | Kev 0.8B | Verdict |
|---|---|---|---|
| Zero-shot IE spectrum: mixed-pool classification accuracy | 83.3 | 72.9 | level |
| Zero-shot IE multilingual: medium-resource-language tier accuracy | 100.0 % | 89.0 % | level |
| Zero-shot IE multilingual: popular-language tier accuracy | 100.0 % | 94.0 % | level |
| Zero-shot IE multilingual: rare-language tier accuracy | 50.0 % | 50.0 % | level |
Source · latest October 17, 2026
All data side by side
| Decider 0.8B | Kev 0.8B | |
|---|---|---|
| Kind | Standalone | Adapter |
| Developer | Mapika | Jared Palmer |
| Status | Open source | Open source |
| Released | n/a | September 20, 2026 |
| Weights | open | open |
| Runs locally | yes | yes |
| Parameters | 0.8 B | n/a |
| License | Apache-2.0 | Apache-2.0 for the adapter and head |
| Popularity (out of 100) | 39 | 45 |